Dear Moses users,
I have a small bilingual corpus, i.e. some tens of thousands of pairs of
phrases, in a quite rare language combination and I would like to see if
this is enough to provide a statistically based translation that is superior
to Google's, also considering that Google appears to use
Hi Barry,
In fact, the directory /home/christian/demo/tools/irstlm/lib/i686 didn't
exist, but /home/christian/demo/tools/irstlm/lib/i386-redhat-linux-gnu did,
and it contained libirstlm.a. After writing ln -s i386-redhat-linux-gnu
i686, everything worked. So in my case the softlinking had to be
Hi,
You don't need to know the total number of the lines in advance. I
attached my modified version of main.c, downsizing the number of
sentences in memory for each iteration to 250. I tested it using
examples/sec??.tagged, and the both input file is parsed without any
segmentation fault.
Best
Hi,
I was working on word-lattice decoding functionality of Moses decoder. In
order to do that, I need to put the morphologically analysed input text into
form of PLF, mentioned on http://www.statmt.org/moses/?n=Moses.WordLattices.
In PLF, each edge includes the information of the produced string
Hello,
Are the order of the phrases in the phrase table same as the sentences corpus?
i mean for example
phrases 0-n extracted from sentences 1
phrases 0-k extracted from sentences 2
phrases 0-j extracted from sentences 3
and so on
--
Best Regards,
S.Bakhshaei
Hi Mehmet,
The following lattice will do what you are asking for, I think:
((('x',0.5,1),('xy',0.5,2)),(('yz',1,2),),(('z',1,1),),)
The trick is ot use the last element of the tuples to indicate what
node the edge ends up in. The first two nodes, have single edges
leaving, but the edges don't
Hi,
I saw that KenLM source code is distributed from the Moses svn and can
set in configure. Is anybody here using it and willing to share some
experiences? Is it thread-safe and can used in Moses together with SRI
and IRST ? Any particular advantages? Is there any more information than
Hello,
I have my own extract and extract.inv,
Is there any way that i make extract.o from these two?
--
Best Regards,
S.Bakhshaei
___
Moses-support mailing list
Moses-support@mit.edu
Hi Moses,
Introducing kenlm in Moses trunk. You no longer need to download a
separate language model to use Moses; it's distributed with Moses and
compiled in by default on UNIX. This is threadsafe language model
inference code that returns the same probabilities as SRI (up to
floating
hi all, can any one give me a litle education on how to build a bilgual corpus
from scratch.tanks kess edor
___
Moses-support mailing list
Moses-support@mit.edu
http://mailman.mit.edu/mailman/listinfo/moses-support
10 matches
Mail list logo